Re: [smartpower-interest] [smartpowerdir] FYI OSTP requests input on SmartGrid

[Andy Bierman <[e-mail]>]

Damir Rajnovic wrote:
> On Sun, Feb 14, 2010 at 12:22:25AM -0800, Fred Baker wrote:
>   
>> Where I have a problem, like you, is the idea that I'm going to spend  
>> my day staring at the electric meter and make random changes at random  
>> times at the whim of the utility. What makes a lot more sense to me is  
>> the idea that the electric utility will give me (or my electronic  
>> proxy) an idea today about pricing issues tomorrow, and I can set up  
>> policies in accordance that intelligent devices in my home will  
>> implement.
>>     
>
> I would say that it must be my electronic proxy. The same way it is unlikely
> for people to watch their meters all day long it is not likely that they
> will review "power pricing forecast" every night and then make adjustments
> to their equipment. We also have to bear in mind that these tarifs and
> times are just forecast. If something significant happens in the grid
> utility may shift times when price is low/high up or down.
>
> So, instead of programming my devices in absolute times, I may want to use
> a mechanism similar that is used for recording TV shows. I can see on TV
> schedule that movie I want to record is after a sport event. But if the
> sport event runs over then the movie will start that later. So, instead
> of programming my recorder for some exact time I tell it to wait for the
> movie code to be sent by the broadcaster and then start recording.
>
> Now substitute "TV schedule" with "power pricing forecast" and "movie"
> with "low tarif" and that is it. I can review if pricing forecast for
> tomorrow/next week is acceptable to me and then decide to utilise it
> or not. Having in mind that price forecast times may change my instruction
> to the proxy may be "use low tarif for heating at all time except 
> between 06:00-07:00".
>
>   

I agree that a person is not going to poll pricing forecasts,
especially to save a few cents on a load of laundry.

However, I do not see how an application can do an
effective job as a proxy, because many household
tasks need to finish once they start, or finish within
a reasonable time window.  (I'm not going to wake up
at 3 am to see if the wet clothes are ready to go into
the dryer.)

It did occur to me that it would be difficult to
schedule low-demand pricing if it was known in
advance, since this would affect the future usage
patterns.  Is the utility going to guarantee that
power will be at rate X from time T1 to T2?
How far in advance can such a request be made?
Unless a 'power reservation' can be made, I
do not see how an application can go a good job
at scheduling.  Threshold-based scheduling
would be prone to fluctuations, and might end up
costing more than on-demand service.


>> BTW, a sudden wide-scale cut-off is bad for the electric utilities  
>> too: imagine them sending out such a command, and within the next  
>> hertz everyone changing their demand levels. Read RFC 3439 section  
>> 2.2.2; the issues and arguments are much the same for the grid as they  
>> are for the Internet.
>>     
>
> Should we then introduce switch off/on interval to spread in time when
> devices will turn off/on? We can give users ability to say "turn off/on
> this device within an interval from <Start> and <Finish> time" and then
> leave the proxy or meter to randomly select exact time when device
> will be turned off/on. The only worry would be for all meters to pick
> the same random number.
>
> Thanks,
>
> Gaus
>
>
>   

Andy

>   
>> But imagine that I was told yesterday (or even an hour in advance  
>> today) that the price would be "low" most of today but between the  
>> hours of <this> and <that> the price would be "high". With 24 hours  
>> notice, I could decide to do my laundry in the morning, for example,  
>> if that was a real issue. With even an hour's notice my home energy  
>> controller could tell my air conditioner and my refrigerator/freezer  
>> to control to a slightly lower temperature for an hour, and then at  
>> the magic time (or five minutes before) it goes back to controlling to  
>> the normal temperature. The refrigerator/freezer, btw, doesn't run the  
>> defrost cycle during the high price interval - it waits until 12 hours  
>> later or so. That has much the same effect of randomization without  
>> either the square wave for the utilities or the stomach acid for me.
>>
>> Alternatively, one could imagine the air conditioner having two  
>> temperature settings, one for when power is cheap and one for when the  
>> cost is high. I'll pay extra to keep the house below 90', perhaps, but  
>> only cool to 78' when power is cheap.
>>
>> Where this fits into the technical work is that the information models  
>> for communications in the AMI (eg between the meter and the utility),  
>> we need some way to have the utility able to say "the price between  
>> <date/time> and <date/time> is <price> per <unit>".
>>
>> It might look like this:
>>
>> <tarrif_event>
>>       <Start ss:Type="DateTime">2010-06-01T13:00:00.000</Start>
>>       <Finish ss:Type="DateTime">2010-06-01T16:59:59</Finish>
>>       <Price ss:Type="Float">0.50</Price>
>> </tarrif_event>
>>
>> be sent from the utility to the home energy controller aand be  
>> interpreted in accordance with the contract with the utility, which  
>> probably specifies the local currency as the unit of measure.
>> _______________________________________________
>> smartpower-interest mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/smartpower-interest
>>     
>
> ==============
> Damir Rajnovic <[e-mail]>om>, PSIRT Incident Manager, Cisco Systems
> <http://www.cisco.com/go/psirt>      Telephone: [phone]
> 200 Longwater Avenue, Green Park, Reading, Berkshire RG2 6GB, GB
> ==============
> There are no insolvable problems. 
> The question is can you accept the solution? 
>
> - - - -
> This e-mail may contain confidential and privileged material for the sole 
> use of the intended recipient. Any review, use, distribution or disclosure by 
> others is strictly prohibited. If you are not the intended recipient (or 
> authorized to receive for the recipient), please contact the sender by reply 
> e-mail and delete all copies of this message.
>
> Cisco Systems Limited (Company Number: 02558939), is registered in England 
> and Wales with its registered office at 1 Callaghan Square, Cardiff, 
> South Glamorgan CF10 5BT
> _______________________________________________
> smartpower-interest mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/smartpower-interest
>